From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754290AbcJIHm0 (ORCPT ); Sun, 9 Oct 2016 03:42:26 -0400 Received: from mga03.intel.com ([134.134.136.65]:16381 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752320AbcJIHmY (ORCPT ); Sun, 9 Oct 2016 03:42:24 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.31,465,1473145200"; d="scan'208";a="1062390971" Date: Sun, 9 Oct 2016 15:41:42 +0800 From: Aaron Lu To: Xin Long Cc: Marcelo Ricardo Leitner , kernel test robot , Stephen Rothwell , lkp@01.org, "David S. Miller" , LKML , "Chen, Tim C" , Huang Ying Subject: Re: [LKP] [lkp] [sctp] a6c2f79287: netperf.Throughput_Mbps -37.2% regression Message-ID: <20161009074142.GA9546@aaronlu.sh.intel.com> References: <80bbef60-34ed-0bf1-f1c5-8285b713c2f7@intel.com> <20160818032156.GA5250@aaronlu.sh.intel.com> <20160819052941.GA1179@aaronlu.sh.intel.com> <20160819072420.GA1167@aaronlu.sh.intel.com> <9ddf610d-c819-14ec-42b0-0ee61c0de173@gmail.com> <2019939b-df57-2b02-287d-fcba0f07c57c@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Oct 03, 2016 at 10:32:04AM +0800, Xin Long wrote: > On Fri, Sep 30, 2016 at 3:05 PM, Aaron Lu wrote: > > On 08/23/2016 05:44 AM, Marcelo Ricardo Leitner wrote: > >> Em 19-08-2016 04:24, Aaron Lu escreveu: > >>> On Fri, Aug 19, 2016 at 04:19:39AM -0300, Marcelo Ricardo Leitner wrote: > >>>> Hi, > >>>> > >>>> Em 19-08-2016 02:29, Aaron Lu escreveu: > >>>> ... > >>>>> It doesn't look insane and sctp_wait_for_sndbuf may actually have > >>>>> something to do with a larger sctp_chunk I suppose? > >>>>> > >>>>> The same perf record doesn't capture any sample for the good commit, > >>>>> which suggests the nerperf process doesn't sleep in sctp_wait_for_sndbuf. > >>>> > >>>> Ahhh yes! It does, and then it would mean your txbuf is too small for the > >>>> chunk sizes you're using (sctp tests option -m). > >>>> > >>>> What's your netperf cmdline again please? > >>> > >>> netperf -4 -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K -H 127.0.0.1 > >>> > >>> Is the 10K used here a problem? If so, can you suggest a proper value > >>> for our netperf performance test? Thanks. > >> > >> We're still working on this. Xin could reproduce it on an i3 too, but > >> I'm afraid this commit just unmasked an issue in there. You're > >> overloading the CPU by too much when spawning 8 parallel netperf's on a > >> 4-core system, seems that commit a6c2f79287 was that last rock that made > >> it slip into a precipice. sctp's cwnd and rwnd management are not as > >> good as tcp's and now it seems you're triggering a corner case. > >> > >> I hope to have more soon. > > > > I wonder if there is any update on this issue? > > > git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git > > be4947b sctp: change to check peer prsctp_capable when using prsctp polices > 0605483 sctp: remove prsctp_param from sctp_chunk > 73dca12 sctp: move sent_count to the memory hole in sctp_chunk > > These three commit can avoid this issue by recovering sctp_chunk size. Thanks for the update, I just confirmed the throughput is back on my desktop. From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============5907024089911636448==" MIME-Version: 1.0 From: Aaron Lu To: lkp@lists.01.org Subject: Re: [sctp] a6c2f79287: netperf.Throughput_Mbps -37.2% regression Date: Sun, 09 Oct 2016 15:41:42 +0800 Message-ID: <20161009074142.GA9546@aaronlu.sh.intel.com> In-Reply-To: List-Id: --===============5907024089911636448== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Mon, Oct 03, 2016 at 10:32:04AM +0800, Xin Long wrote: > On Fri, Sep 30, 2016 at 3:05 PM, Aaron Lu wrote: > > On 08/23/2016 05:44 AM, Marcelo Ricardo Leitner wrote: > >> Em 19-08-2016 04:24, Aaron Lu escreveu: > >>> On Fri, Aug 19, 2016 at 04:19:39AM -0300, Marcelo Ricardo Leitner wro= te: > >>>> Hi, > >>>> > >>>> Em 19-08-2016 02:29, Aaron Lu escreveu: > >>>> ... > >>>>> It doesn't look insane and sctp_wait_for_sndbuf may actually have > >>>>> something to do with a larger sctp_chunk I suppose? > >>>>> > >>>>> The same perf record doesn't capture any sample for the good commit, > >>>>> which suggests the nerperf process doesn't sleep in sctp_wait_for_s= ndbuf. > >>>> > >>>> Ahhh yes! It does, and then it would mean your txbuf is too small fo= r the > >>>> chunk sizes you're using (sctp tests option -m). > >>>> > >>>> What's your netperf cmdline again please? > >>> > >>> netperf -4 -t SCTP_STREAM_MANY -c -C -l 300 -- -m 10K -H 127.0.0.1 > >>> > >>> Is the 10K used here a problem? If so, can you suggest a proper value > >>> for our netperf performance test? Thanks. > >> > >> We're still working on this. Xin could reproduce it on an i3 too, but > >> I'm afraid this commit just unmasked an issue in there. You're > >> overloading the CPU by too much when spawning 8 parallel netperf's on a > >> 4-core system, seems that commit a6c2f79287 was that last rock that ma= de > >> it slip into a precipice. sctp's cwnd and rwnd management are not as > >> good as tcp's and now it seems you're triggering a corner case. > >> > >> I hope to have more soon. > > > > I wonder if there is any update on this issue? > > > git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git > = > be4947b sctp: change to check peer prsctp_capable when using prsctp polic= es > 0605483 sctp: remove prsctp_param from sctp_chunk > 73dca12 sctp: move sent_count to the memory hole in sctp_chunk > = > These three commit can avoid this issue by recovering sctp_chunk size. Thanks for the update, I just confirmed the throughput is back on my desktop. --===============5907024089911636448==--